Re: Concensus Call on Urgent Pointer.

[Vern Paxson <vern@ee.lbl.gov>]

> 5/Vern - the formula you had in your email is interesting - Can you give me
> a pointer to its derivation.  I would also like to see formulas which had
> the line rate and window size as parameters.  The formula in your email
> aims to obtain a theoretical max throughput.  This is interesting. However
> on a Gigabit line if the error rate is 0 the throughput is not infinite ,
> it is 1Gbps.

The basic formula is in:

	Matthew Mathis, Jeffrey Semke, Jamshid Mahdavi, and Teunis Ott,
	"The Macroscopic Behavior of the Congestion Avoidance Algorithm",
	Computer Communications Review, volume 27, number 3, July 1997.

	http://www.psc.edu/networking/papers/model_abstract.html

A more refined version is in:

	Modeling TCP Throughput: A Simple Model and its Empirical Validation,
	Jitendra Padhye, Victor Firoiu, Don Towsley, and Jim Kurose (U. Mass),
	Proc. SIGCOMM '98.

	http://www.acm.org/sigcomm/sigcomm98/tp/abs_25.html

(This version takes into account the effect of timeouts.)

I quoted the short form of the formula.  The more detailed versions account
for the line rate and window size (these basically just introduce upper bounds
on the achievable throughput).

> Also, on a Gbps line if the packet error rate is 10E-6 and
> the window is of infinite size then the throughput is not affected much.

The formula gives an upper bound (for average sustained throughput) even if
the window is of infinite size, so that error rate will still limit the
throughput accordingly.

		Vern